README.md

In deruncie/GridLMM: Efficient Mixed Models for GWAS with multiple Random Effects

GridLMM is a package for fitting linear mixed models (LMMs) with multiple random effects.

The fitting process is optimized for repeated evaluation of the random effect model with different sets of fixed effects, ex. for GWAS analyses. The approximation is due to the use of a discrete grid of possible values for the random effect variance component proportions. We include functions for both frequentist and Bayesian GWAS, (Restricted) Maximum Likelihood evaluation and Bayesian Posterior inference of variance components.

Please treat this as a Beta version and let me know of issues running the functions.

Installation:

devtools::install_github('deruncie/GridLMM')

Main functions:

• GridLMM_ML: estimates parameters of a LMM by (restricted) Maximum Likelihood
• GridLMM_posterior: Approximates the posterior distribution of the variance component proportions of a LMM
• GridLMM_GWAS: Runs a GWAS with the error structure a LMM. By default, uses heuristics to efficiently sample the grid. Can run Wald tests (method = 'REML'), Likelihood ratio tests (method = 'ML'), or calculate Bayes Factors (method = 'BF')

Examples:

There is a vignette walking through the data format necessary for GridLMM and a few analyses using GridLMM_GWAS.

If you would like to build the vignette (see below), do:

devtools::install_github('deruncie/GridLMM', build_opts = c("--no-resave-data", "--no-manual"),force = TRUE,build_vignettes = TRUE)

vignette(topic = 'Running_GridLMM_GWAS',package='GridLMM')

deruncie/GridLMM documentation built on May 2, 2023, 7:18 p.m.